LINKING TRACKED OBJECTS THAT UNDERGO TEMPORARY OCCLUSION

• US 20080226127A1
• Filed: 01/22/2008
• Published: 09/18/2008
• Est. Priority Date: 02/10/2003
• Status: Active Grant

• Current Assignee: Tomas Brodsky, Yun-Ting Lin
• Original Assignee: Tomas Brodsky, Yun-Ting Lin
• Inventors: Lin, Yun-Ting, Brodsky, Tomas
• Granted Patent: US 7,620,207 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06V 10/24 Aligning, centring, orientation detection or correction of the image

Abstract

A method and system is configured to characterize regions of an environment by the likelihoods of transition of a target from each region to another. The likelihoods of transition between regions is preferably used in combination with conventional object-tracking algorithms to determine the likelihood that a newly-appearing object in a scene corresponds to a recently-disappeared target. The likelihoods of transition may be predefined based on the particular environment, or may be determined based on prior appearances and disappearances in the environment, or a combination of both. The likelihoods of transition may also vary as a function of the time of day, day of the week, and other factors that may affect the likelihoods of transitions between regions in the particular surveillance environment.

Claims

1. A method comprising:

displaying a first image corresponding to a field of view of at least one camera,

providing a user interface that allows a user to;

identify a first region of the first image at which objects may disappear from the image,

identify a second region of a second image at which objects may appear in the other image, and

define a first likelihood that an object that disappears from the first region will appear in the second region, and

subsequently identifying a target object in the first image, and determining whether the target object appears in the second image based on the first likelihood and an object similarity measure.

2. The method of claim 1, wherein the first region of the first image corresponds to a doorway.

3. The method of claim 1, wherein the first image corresponds to section of a floorplan, and the second image corresponds to an other section of the floorplan.

4. The method of claim 1, wherein the likelihood is defined as a function of at least one of:

a time of day,

a day of a week, and

a season of a year.

5. The method of claim 1, including updating the likelihood based on occurrences of objects disappearing from the first region and appearing in the second region.

6. The method of claim 1, including:

allowing the user to;

identify a third region in the first image at which objects may appear in the first image, and

define a second likelihood that the object that disappears from the first region will appear in the third region, and

determining whether the target object reappears in the first image based on the second likelihood and an other object similarity measure.

7. The method of claim 1, including allowing the user to define a second likelihood that an object that disappears from the second region will appear in the first region.

8. The method of claim 1, wherein defining the first likelihood includes accepting a default likelihood as the first likelihood.

9. The method of claim 1, wherein identifying the first region includes using a screen cursor to define a bounding rectangle.

10. The method of claim 1, including modifying the first likelihood based on occurrences of transitions between the first region and the second region, wherein the modifying includes applying a low-pass filter to combine a frequency of the occurrences to the first likelihood.

11. The method of claim 1, including allowing the user to:

assign a first name to the first region and a second name to the second region, and define the first likelihood via a reference to the first and second names.

12. A method comprising:

partitioning a camera image into a set of regions,

defining a set of likelihoods of transition of objects between pairs of regions, and

tracking target objects that traverse the camera image based on the set of likelihoods of transition.

13. The method of claim 12, wherein defining the set of likelihoods includes defining a higher likelihood of transition between horizontally neighboring regions than between vertically neighboring regions.

14. The method of claim 12, including modifying one or more of the likelihoods based on occurrences of objects transitioning between pairs of regions.

15. A system comprising:

a display processor that is configured to display a first image corresponding to a field of view of at least one camera,

a user interface that is configured to allow a user to;

identify a first region of the first image at which objects may disappear from the image,

identify a second region of a second image at which objects may appear in the other image, and

define a first likelihood that an object that disappears from the first region will appear in the second region, and

an object comparer that is configured to;

compare a target object in the first image to an other object in the second image to provide an object similarity measure, and

determine whether the other object in the second image corresponds to the target object, based on the first likelihood and the object similarity measure.

16. The system of claim 15, wherein the first region of the first image corresponds to a doorway.

17. The system of claim 15, wherein the first image corresponds to section of a floorplan, and the second image corresponds to an other section of the floorplan.

18. The system of claim 15, wherein:

the user interface is configured to allow the user to;

identify a third region in the first image at which objects may appear in the first image, and

define a second likelihood that the object that disappears from the first region will appear in the third region, and

the object comparer is configured to determine whether the target object reappears in the first image based on the second likelihood and an other object similarity measure.

19. The system of claim 15, wherein the user interface is configured to allow the user to identify the first region by using a screen cursor to define a bounding rectangle.

20. The system of claim 15, including a transition predictor that includes a low-pass filter that is configured to modify the first likelihood based on occurrences of transitions between the first region and the second region.